<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Drops on Suraj Tripathi</title><link>https://surajtripathi.info/drops/</link><description>Recent content in Drops on Suraj Tripathi</description><generator>Hugo</generator><language>en</language><lastBuildDate>Fri, 30 Oct 2026 21:00:00 +0530</lastBuildDate><atom:link href="https://surajtripathi.info/drops/index.xml" rel="self" type="application/rss+xml"/><item><title>A retry job that retired good rows</title><link>https://surajtripathi.info/drops/2026-10-30/</link><pubDate>Fri, 30 Oct 2026 21:00:00 +0530</pubDate><guid>https://surajtripathi.info/drops/2026-10-30/</guid><description>&lt;p&gt;&lt;img src="cover.png" alt=""&gt;&lt;/p&gt;
&lt;p&gt;&lt;a href="https://github.com/surajtripathy07/membral" class="external-link" target="_blank" rel="noopener"&gt;Membral&lt;/a&gt;, a tool I build, saves your AI coding chat history so you can search it later. To make that history searchable, a small model running on my own machine turns each saved chunk of text, stored as one row in a database, into a vector, a list of numbers that captures what the text means.&lt;/p&gt;
&lt;p&gt;When that step fails, a retry job tries the chunk again later, but in review I found it gave up on a chunk forever after a single failure, even when the model server was only briefly down. The server had just returned a 500 error, a generic &amp;ldquo;something broke on my side&amp;rdquo; reply, on a batch of 32 chunks, so one bad minute would have left 32 good rows unsearchable for good.&lt;/p&gt;</description></item></channel></rss>